Research On Pressure Straightening Strategy And Intelligent Control Technology For LSAW Pipes

The principal type of longitudinally submerged arc welding (LSAW) pipes used as oiland gas pipelines, is often formed through two methods in the world: UOE (U-FormingO-Forming and Expanding) process and JCOE (J-Forming C-Forming O-Forming andExpanding) process. In the LSAW pipe manufacturing process, there often arises ademand to straighten the end of LSAW pipes because the pipes fail to meet the overallstraightness requirements, due to the influence of such factors as the welding thermalstress, the forming equipment, and the overall straightness of the mould. According to thegeometric particularity of the LSAW pipes, the pressure straightening method tends to beapplied at present by many manufacturers to correct straightness of the pipes. Due tolimitations on the current three-point bending pressure straightening process, one singlestep straightening can only straighten the pipe from a big single-curvature curve into asmall double-curvature curve in the shape of the letter “S”. One single step strengtheningwould work efficiently for pipes of which the initial deflection relatively small, but whenthe initial deflection of a pipe is big enough, the multi-step straightening process would berequired to achieve the goal for the improved overall straightness on the pipe. Since thedeflection distribution over the pipe length needs to be measured prior to each single stepstraightening in order to determine the parameters such as the load position, the supportlocation, and the straightening stroke, etc., the multi-step straightening process is too slowto satisfy the requirements of the industry.To overcome the deficiencies on the existing models and solve the above discussedproblems, a quantitative method is firstly established to calculate the theoreticalstraightening moment according to a pipe’s initial deflection distribution, which is basedon the springback equation of small curvature plane bending previously developed.Theoretically, the pipe could be completely straightened in one go when such astraightening moment is imposed. As a theoretical basis, quantification of the theoreticalstraightening moment distribution makes it possible to study various pressurestraightening control strategies. To facilitate industrial applications, the new multi-point bending one-offstraightening control strategy is then developed, along with the method for obtainingcorresponding straightening parameters by discretizing and linearizing the theoreticalmoment curve. The Finite Element Methods (FEM) simulation results from the LSAWpipe indicates: the more number of pressure points, the higher the straightening accuracy.Furthermore, the FEM simulation model is validated by physical simulation experimentsof small sized pipes. To reduce the number of pressure points and ensure the straighteningaccuracy, the concept of the load correction coefficient is introduced and the methods toobtain an optimal correction coefficient are then excessively studied: least squares curvefitting method and equivalent bending deformation energy method.The FEM simulationresults indicates that the optimum load correction coefficient, which is calculated by theequivalent bending deformation energy method, is more applicable. Feasibility andreliability of the newly developed control strategy is verified through a FEM simulationmodel, and it provides a way to quantitatively implement the multi-point bending one-offstraightening control strategy.For the current three-point bending pressure straightening process, the principle ofthe multi-step pressure straightening process is firstly revealed: a jagged polylinedistribution moment is applied to approach the smooth curve of the theoreticalstraightening moment. Therefore, the quantitative control method of the multi-stepthree-point bending pressure straightening is proposed. In this method, the straighteningcontrol parameters used in each three-point bending step in the multi-step pressurestraightening can be obtained by measuring the initial deflection distribution of the pipeonce only. Feasibility and reliability of the quantitative control method are verified byphysical simulation experiments of the small sized pipes with different geometries,different initial deflection distributions and different straightening steps. It not onlyensures the straightening accuracy, and greatly improves the straightening efficiency.In view of the deflection characteristics of LSAW pipes, a self-propelled straightnessinspection system is designed. In this system, using a remote control car as a carrier of thelaser displacement sensor, a pipe is measured automatically on the orbit of frame type.With LabVIEW virtual instrument as software platform, system measurement routine is used to achieve data collection, output and display. Accuracy and reliability of theself-propelled straightness inspection system are verified by the measurement experimentsof the small sized pipes with different deflections.Accurately identifying the material parameters of pipes is an important condition ofimproving straightening precision. The material parameters always have large fluctuationsbecause of many factors， such as the material batch, heat treatment, the deformationhistory, etc. In this paper, firstly, the approximate relation between the small curvaturepipe in pressure straightening process and the straight pipe in three-point bending process,whose cross section and material are same with the former, has been verified by finiteelement method. Based on the theoretical analysis for three-point bending process, thispaper proposes the online recognition model for material parameters with the neuralnetwork method. The feasibility and the reliability of the online recognition model areverified by the experiments of the small sized pipes, which can be directly applied to thecalculation of pressure straightening process parameters.According to the straightness inspection system and material performance parameterson-line identification model, the specific implementation steps of intelligent multi-pointbending straightening control strategy and intelligent multi-step three-point bendingstraightening control strategy are given respectively that provides the theoretical principleto build the intelligent straightening control system in the future.